Evidence review for monitoring for gastric cancer in people with vitamin B12 deficiency due to autoimmune gastritis

Evidence review for monitoring for gastric cancer in people with vitamin B12 deficiency due to autoimmune gastritis

Vitamin B12 deficiency in over 16s: diagnosis and management

Evidence review G

NICE Guideline, No. 239

London: National Institute for Health and Care Excellence (NICE); 2024 Mar.

ISBN-13: 978-1-4731-5735-4

1. Monitoring for gastric cancer in people with vitamin B12 deficiency due to autoimmune gastritis

1.1. Review question

What monitoring should be offered to people with pernicious anaemia to identify gastric cancer?

1.1.1. Introduction

Intrinsic factor is essential for the normal absorption of vitamin B12. Intrinsic factor is produced by the parietal cells of the stomach. Inflammatory conditions of the stomach, predominantly autoimmune gastritis and less commonly H. pylori infection, can lead to the dysfunction and the eventual loss of parietal cells in the stomach, resulting in vitamin B12 deficiency. Previous studies have suggested that several types of cancer affecting the cells of the stomach may be more likely to occur in some people who have vitamin B12 deficiency. This increased risk of stomach cancer development is not due to the vitamin B12 deficiency itself, but the associated stomach inflammation. The two main types of stomach cancer involved are gastric adenocarcinomas and gastric neuroendocrine tumours. The increased risk is therefore only found in those patients who have a gastric cause of vitamin B12 deficiency. This association is plausible given that both the chronic inflammation and the physiological changes such as reduced gastric acid secretion are possible risk factors for stomach cancer development.

The overall risk of cancers affecting the stomach in people with autoimmune gastritis is uncertain. There is no standardised approach to screening and surveillance of the stomach for cancers. The yield and clinical benefit of such screening, as well as the optimal methodology and frequency of monitoring, have not been previously defined.

This review seeks to assess the most clinically and cost-effective monitoring strategy to identify gastric cancer in people with vitamin B12 deficiency due to autoimmune gastritis, including the type of procedures and frequency.

1.1.2. Summary of the protocol

For full details see the review protocol in Appendix A.

Table 1

PICO characteristics of review question.

1.1.3. Methods and process

This evidence review was developed using the methods and process described in Developing NICE guidelines: the manual. Methods specific to this review question are described in the review protocol in appendix A and the methods document.

Declarations of interest were recorded according to NICE’s conflicts of interest policy.

1.1.4. Effectiveness evidence

1.1.4.1. Included studies

A search was conducted for randomised trials comparing the effectiveness of monitoring strategies with each other or with no monitoring for gastric cancer in people with pernicious anaemia. No randomised trials were identified that included people with pernicious anaemia only. One randomised controlled trial² comparing gastroscopic follow up at 24 months with gastroscopic follow up at 48 months in people with atrophic gastritis was identified. 59.5% of study participants had pernicious anaemia and were therefore considered an indirect population and evidence was downgraded. The study is summarised in Table 2 below. Evidence from this study is summarised in the clinical evidence summary below (Table 4).

As no direct evidence from randomised trials was identified, a search was conducted for non-randomised studies. Two cohort studies¹^, ⁶ comparing gastroscopy with no monitoring were included. The studies are summarised in Table 3 below. Evidence from these studies is summarised in the clinical evidence summary below (Table 5 and 6).

None of the included studies reported whether participants had undergone previous gastric surgery. Evidence was identified for mortality, incidence of carcinoid tumours and gastric carcinoma. No evidence was identified for quality of life, stage of cancer at diagnosis/surgical resectability or adverse events.

No evidence was identified for the effectiveness of barium meal, pepsinogen (followed by gastroscopy for those at high risk), gastrin (followed by gastroscopy for those at high risk), combined pepsinogen + gastrin (followed by gastroscopy for those at high risk), 3 staged pepsinogen, followed by gastrin (followed by gastroscopy for those at high risk), or 3 staged gastrin, followed by pepsinogen (followed by gastroscopy for those at high risk).

See also the study selection flow chart in Appendix C, study evidence tables in Appendix D, forest plots in Appendix E and GRADE tables in Appendix F.

1.1.4.2. Excluded studies

See the excluded studies list in Appendix J.

1.1.5. Summary of studies included in the effectiveness evidence

Table 2

Summary of randomised controlled trials included in the evidence review.

Table 3

Summary of non-randomised studies included in the evidence review.

See Appendix D for full evidence tables.

1.1.6. Summary of the effectiveness evidence

Table 4

Clinical evidence summary: Gastroscopy at 24 months versus gastroscopy at 48 months (RCT evidence).

Table 5

Clinical evidence summary: Gastroscopy (mean every 14 months) versus no monitoring.

Table 6

Clinical evidence summary: Gastroscopic screening versus no monitoring.

See Appendix F for full GRADE tables.

1.1.7. Economic evidence

1.1.7.1. Included studies

One health economic study with relevant comparisons was included in this review.³ The study compared using monitoring for cancer using gastroscopy in different population groups. This is summarised in the health economic evidence profile below (Table 7) and the health.

The patient population in the included study all had underlying atrophic gastritis, pernicious anaemia was a stratum in the study.

1.1.7.2. Excluded studies

No relevant health economic studies were excluded due to assessment of limited applicability or methodological limitations.

See also the health economic study selection flow chart in Appendix G.

1.1.8. Summary of included economic evidence

Table 7

Health economic evidence profile: Gastroscopy surveillance for people with pernicious anaemia.

1.1.9. The committee’s discussion and interpretation of the evidence

1.1.9.1. The outcomes that matter most

The committee considered mortality, quality of life, diagnosis of cancer, stage of cancer at diagnosis or surgical resectability, incidence of gastric neuroendocrine tumours and procedure related adverse events including bleeding, perforation, and aspiration to be the most important outcomes of monitoring for gastric cancer. All outcomes were considered equally important for decision making and therefore were all rated as critical.

The purpose of the outcome of stage of cancer at diagnosis or surgical resectability was to determine whether the monitoring strategies help to identify cancer earlier, that is, before it progresses to a higher stage, or beyond surgical removal. However, no evidence was identified for this outcome. No evidence was identified for quality of life, or procedure related adverse events.

1.1.9.2. The quality of the evidence

Evidence came from one randomised controlled trial comparing different lengths of gastroscopic follow up and two observational cohort studies comparing gastroscopic follow up with no monitoring. The quality of the evidence for all identified outcomes was very low. The main reasons for downgrading of the quality of the evidence were indirectness, risk of bias and imprecision. No evidence was identified for monitoring with barium meal, pepsinogen, gastrin, or any combination/staged protocols using pepsinogen and gastrin.

One of the main reasons for downgrading the quality of the evidence was population indirectness. The committee decided when setting the review protocol to stratify the evidence for people who have undergone previous gastric surgery, as the associated risk of developing gastric cancer is higher in this group. However, none of the included studies reported this information, therefore all outcomes were downgraded for serious population indirectness. In addition, the study population in the randomised controlled trial was people with body-predominant atrophic gastritis, sixty per cent of whom had pernicious anaemia. As forty per cent of the study population did not have pernicious anaemia, the evidence from this study was downgraded further for population indirectness. The committee considered the evidence to be relevant to this review as people with pernicious anaemia have atrophic gastritis and it is the atrophic gastritis that increases the risk of gastric cancer rather than the pernicious anaemia itself.

The committee noted the age of the studies, particularly those conducted over thirty years ago. The committee considered that endoscopic techniques have greatly improved since the studies were carried out and more detailed imaging is now available, allowing detection of smaller or more subtle abnormalities. Therefore, the evidence identified may underestimate the effectiveness of gastroscopic monitoring.

All evidence was at high or very high risk of bias. In the randomised controlled trial, this was due to the lack of information reported on the randomisation process, deviations from the intended interventions and missing outcome data. In the observational studies, this was due to the lack of adjustment for confounders, classification of interventions and missing data.

The included studies all had small samples sizes, around one hundred participants or less. This led to imprecision around several of the point estimates. The committee were aware that around one in three hundred people with autoimmune gastritis per year develop gastric cancer. Therefore, the studies were likely to be underpowered to detect any meaningful differences between the monitoring strategies.

Considering all the limitations outlined above, the committee considered there to be insufficient evidence upon which to base recommendations. The committee agreed that this is an area in which further research is needed and therefore decided to make a recommendation for research.

1.1.9.3. Benefits and harms

All the evidence identified was for gastroscopy, with no evidence identified for any other monitoring technique listed in the review protocol. The committee noted that gastroscopy is the only technique that can detect the presence of gastric cancer in people with symptoms. The other methods are more appropriate in the context of assessing risk of developing cancer in people who are asymptomatic.

Very low-quality evidence suggested a benefit of regular gastroscopy over no monitoring for reducing mortality, although there was very serious imprecision, with wide confidence intervals compatible with no difference and a harm of gastroscopy. Very low-quality evidence also showed a benefit of gastroscopic screening over no monitoring for increasing the identification of carcinoid tumours. No clinically important difference was found between twenty-four and forty-eight month follow up with gastroscopy in the identification of carcinoid tumours. No clinically important difference was found between regular gastroscopy or gastroscopic screening compared with no monitoring for gastric carcinoma. The committee considered there to be insufficient evidence upon which to base recommendations.

The committee considered what is being done in current clinical practice. Gastroscopy on diagnosis of autoimmune gastritis is not carried out routinely and referral to secondary care depends on local interest and locally available services, GP knowledge and awareness, and tradition. The committee also considered the potential harms of gastroscopic screening or monitoring. No evidence was identified for quality of life or procedure related adverse events; however, the committee discussed the physically uncomfortable nature of the procedure from the patients’ perspective. Therefore, the committee agreed that without sufficient evidence to recommend gastroscopic surveillance for gastric cancer in all people with autoimmune gastritis, gastroscopy should only be offered if there is a clinical reason for doing so.

If there are symptoms of gastric cancer present, such as new dysphagia, or dyspepsia, nausea and vomiting and weight loss in over 55’s, people are referred urgently to gastroenterology through the two-week cancer referral pathway. Biopsies are taken at gastroscopy and only those with high operative link for gastritis assessment (OLGA) and operative link for gastric intestinal metaplasia assessment (OLGIM) scores would have subsequent or regular gastroscopic monitoring. Identification of small tumours (<10mm) for example, require surveillance rather than treatment. The committee recommended that healthcare professionals consider referral for gastroscopic endoscopy if the person has autoimmune gastritis with new onset upper gastrointestinal symptoms.

The committee highlighted the need for greater awareness on the upper gastrointestinal symptoms, as well as the increased risk of gastric adenocarcinoma and gastric neuroendocrine tumours in people with autoimmune gastritis. Therefore, this information was included in the wording of the recommendations in the hope that this will raise awareness, particularly among GPs. The committee were also aware of the NICE guideline on Suspected cancer and cross referred to the recommendations on lower and upper gastrointestinal tract cancers.

1.1.9.4. Cost effectiveness and resource use

Economic evidence

One economic evaluation was identified for this review. The economic evaluation investigated the cost to detect cancer for people with autoimmune gastritis compared to people with additional risk factors with autoimmune gastritis using gastroscopy surveillance. The patient population in the included study all had underlying atrophic gastritis and autoimmune gastritis was a stratum in the study. The committee considered only the results for people with autoimmune gastritis.

In the economic evaluation, the cost to detect a cancer for an autoimmune gastritis patient without extensive atrophic gastritis was £14,486 whilst the cost to detect a cancer in an autoimmune gastritis patient population with extensive atrophic gastritis was £2,692.

The only reported cost was the cost of the gastroscopy (endoscopy) which was valued at £155. There were no costs of cancer treatment included. Routine gastroscopy was not recommended due to the limitations of the study which didn’t consider quality-adjusted life-years or the cost of cancer treatment. Also, the reported gastroscopy cost was deemed to be significantly lower than other published costs; there is a cost of £754 reported in the ‘National schedule of NHS costs 2020-2021’ which would raise the cost to detect gastric cancer to approximately £70,000 for people without extensive atrophic gastritis. This would then potentially have a significant resource impact on the NHS. Furthermore, there was a relatively small sample of people with autoimmune gastritis (102 people) which added to the uncertainty of the results. This was also a non-UK based study which raised applicability concerns.

No cost-effectiveness evidence was identified for other interventions for monitoring of gastric cancer in people with autoimmune gastritis. There was no evidence for the use of barium meal, pepsinogen, gastrin, or any combination/staged protocols using pepsinogen and gastrin for monitoring.

Potential for modelling

The committee considered whether it would be feasible to model the cost-effectiveness of gastroscopy for surveillance of gastric cancer for people with autoimmune gastritis. The committee thought that earlier treatment could lead to better outcomes and care savings but there is not the evidence to quantify these outcomes.

The committee noted that gastroscopy was not always preferred by people due to the uncomfortable nature of the intervention which may impact their utility whilst undergoing the screening. From the clinical review, there was no evidence relating to quality of life.

The committee were concerned about the lack of available clinical data to inform an economic model specifically for people with autoimmune gastritis and the outcomes for people with autoimmune gastritis that are diagnosed with gastric cancer. The committee were unaware of data relating to the incidence of gastric cancer in people with autoimmune gastritis and the type of cancer as well as stage of cancer when detected. The committee considered whether evidence from other clinical conditions could inform modelling, but they were concerned about the clinical validity of extrapolating evidence from other clinical conditions.

Conclusions about cost effectiveness

There is a substantial cost of offering gastroscopic surveillance of £754 every two-four years. Earlier identification may improve treatment outcomes and patient QALYs however, there is too much uncertainty and lack of evidence to indicate whether the cost-effectiveness of providing routine gastroscopy is less than £20,000 per QALY gained.

There was no evidence at all for other forms of surveillance.

Recommendations

Due to insufficient clinical and cost effectiveness evidence, recommendations for gastroscopy to be offered routinely for surveillance could not be made. There was no evidence at all for other forms of surveillance. The committee members decided to recommend investigating what monitoring should be offered to people with autoimmune gastritis to identify gastric cancer as a research question, and this topic was not subsequently modelled.

The committee also recommended that clinicians look out for symptoms that might suggest cancer and cross-referred to relevant NICE guidance on cancer diagnosis.

Resource impact

The committee members thought that some people with autoimmune gastritis that have gastrointestinal symptoms may already be receiving gastroscopic surveillance. Although there was not enough evidence to recommend routine surveillance, the committee did not expect surveillance to stop and so it is not thought there will be any additional resource impact.

1.1.10. Recommendations supported by this evidence review

This evidence review supports recommendations 1.7.1 and 1.7.2, and the research recommendation on what monitoring should be offered to people with autoimmune gastritis (also known as pernicious anaemia) to identify gastric cancer.

1.1.11. References

1.: Armbrecht U, Stockbrugger RW, Rode J, Menon GG, Cotton PB. Development of gastric dysplasia in pernicious anaemia: a clinical and endoscopic follow up study of 80 patients. Gut. 1990; 31(10):1105–1109 [PMC free article: PMC1378732] [PubMed: 2083855]

2.: Lahner E, Caruana P, D’Ambra G, Ferraro G, Di Giulio E, Delle Fave G et al. First endoscopic-histologic follow-up in patients with body-predominant atrophic gastritis: when should it be done? Gastrointestinal Endoscopy. 2001; 53(4):443–448 [PubMed: 11275884]

3.: Lahner E, Hassan C, Esposito G, Carabotti M, Zullo A, Dinis-Ribeiro M et al. Cost of detecting gastric neoplasia by surveillance endoscopy in atrophic gastritis in Italy: A low risk country. Digestive and liver disease : official journal of the Italian Society of Gastroenterology and the Italian Association for the Study of the Liver. 2017; 49(3):291–296 [PubMed: 28034664]

4.: National Institute for Health and Care Excellence. Developing NICE guidelines: the manual [updated January 2022]. London. National Institute for Health and Care Excellence, 2014. Available from: http://www.nice.org.uk/article/PMG20/chapter/1%20Introduction%20and%20overview

5.: Organisation for Economic Co-operation and Development (OECD). Purchasing power parities (PPP). 2021. Available from: https://data.oecd.org/conversion/purchasing-power-parities-ppp.htm Last accessed: 01/06/2022.

6.: Sjoblom SM, Sipponen P, Miettinen M, Karonen SL, Jrvinen HJ. Gastroscopic screening for gastric carcinoids and carcinoma in pernicious anemia. Endoscopy. 1988; 20(2):52–56 [PubMed: 3383791]

Appendices

Appendix A. Review protocols

Review protocol for monitoring for gastric cancer in people with pernicious anaemia (PDF, 168K)

Health economic review protocol (PDF, 135K)

Appendix B. Literature search strategies

The literature searches for this review are detailed below and complied with the methodology outlined in Developing NICE guidelines: the manual.⁴

For more information, please see the Methodology review published as part of the accompanying documents for this guideline.

B.1. What monitoring should be offered to people with pernicious anaemia to identify gastric cancer?

B.1.1. Clinical search literature search strategy (PDF, 126K)

B.1.2. Health Economics literature search strategy (PDF, 135K)

Appendix I. Health economic model

None.

Appendix J. Excluded studies

Clinical studies

Table 13

Studies excluded from the clinical review.

Health Economic studies

None.

Appendix K. Research recommendations – full details

K.1. Research recommendation

What monitoring should be offered to people with autoimmune gastritis (also known as pernicious anaemia) to identify gastric cancer?

K.1.1. Why this is important

People with autoimmune gastritis have and an increased risk of gastric adenocarcinoma and gastric neuroendocrine tumours compared with the general population. There is insufficient evidence to determine the type and extent of monitoring for gastric cancer that leads to the best outcomes for people with autoimmune gastritis.

K.1.2. Rationale for research recommendation

Download PDF (94K)

K.1.3. Modified PICO table

Download PDF (115K)

Final

Evidence reviews underpinning recommendations 1.7.1 and 1.7.2 and the recommendation for research in the NICE guideline

Developed by NICE

Disclaimer: The recommendations in this guideline represent the view of NICE, arrived at after careful consideration of the evidence available. When exercising their judgement, professionals are expected to take this guideline fully into account, alongside the individual needs, preferences and values of their patients or service users. The recommendations in this guideline are not mandatory and the guideline does not override the responsibility of healthcare professionals to make decisions appropriate to the circumstances of the individual patient, in consultation with the patient and/or their carer or guardian.

Local commissioners and/or providers have a responsibility to enable the guideline to be applied when individual health professionals and their patients or service users wish to use it. They should do so in the context of local and national priorities for funding and developing services, and in light of their duties to have due regard to the need to eliminate unlawful discrimination, to advance equality of opportunity and to reduce health inequalities. Nothing in this guideline should be interpreted in a way that would be inconsistent with compliance with those duties.

NICE guidelines cover health and care in England. Decisions on how they apply in other UK countries are made by ministers in the Welsh Government, Scottish Government, and Northern Ireland Executive. All NICE guidance is subject to regular review and may be updated or withdrawn.

Bookshelf ID: NBK603289PMID: 38691633

Table 1PICO characteristics of review question

Population	Inclusion: adults with diagnosed pernicious anaemia (diagnosis as defined by the studies) Exclusion: other types/causes of vitamin B12 deficiency Strata: Other risk factors for gastric cancer (patients with any previous gastric surgery, including bariatric surgery)
Interventions	Monitoring for gastric cancer: Gastroscopy Barium meal Pepsinogen (followed by gastroscopy for those at high risk) Gastrin (followed by gastroscopy for those at high risk) Combined pepsinogen + gastrin (followed by gastroscopy for those at high risk) 3 staged pepsinogen, followed by gastrin (followed by gastroscopy for those at high risk) 3 staged gastrin, followed by pepsinogen (followed by gastroscopy for those at high risk) Stratify by: Frequency
Comparisons	All monitoring strategies compared with each other (what is included and frequency of monitoring) No monitoring
Outcomes	All outcomes are considered equally important for decision making and therefore have all been rated as critical: Mortality Quality of life Diagnosis of cancer Stage of cancer at diagnosis/surgical resectability Incidence of gastric neuroendocrine tumours (AKA carcinoid tumours/NETS/NENS) Adverse events (procedure related): bleeding perforation aspiration
Study design	Randomised controlled trials Systematic reviews of RCTs Non-randomised studies if insufficient RCT evidence is identified (priority will be given to inclusion of non-randomised comparative studies that have controlled/adjusted for confounding factors. If insufficient evidence is identified from studies that have controlled/adjusted for confounding factors, non-randomised comparative studies that have not controlled/adjusted for confounding factors will be considered) Key confounders: previous gastric surgery Published NMAs and IPDs will be considered for inclusion.

Table 2Summary of randomised controlled trials included in the evidence review

Study

Intervention and comparison

Population

Outcomes

Comments

Lahner 2001²

Gastroscopy at 24 months N=30

Invited to undergo follow-up at a median of 24 months (range 20-26 months) after diagnosis of body-predominant atrophic gastritis

Versus

Gastroscopy at 48 months N=31

Follow-up scheduled at a median of 48 months (range 38-60 months)

Outpatients with body-predominant atrophic gastritis recruited in a screening program for early detection of BAG in patients with unexplained anaemia or long-standing dyspepsia. 59.5% had pernicious anaemia.

Population strata (previous gastric surgery): not reported

Age: group A median (range) 55.5 (22-68) years, group B 59.5 (28-73) years

Sex: 15 males, 27 females

At 2 vs. 4 years:

Carcinoid tumours

Criteria for diagnosis of body–predominant atrophic gastritis: fasting gastrin above upper normal values, hypochlorhydria/achlorhydria to pentagastrin stimulation, and histologic confirmation of gastric body mucosal atrophy.

Criteria for diagnosis of pernicious anaemia: macrocytic anaemia (haemoglobin levels less than 14 g/dL for men and less than 12 g/dL for women; mean corpuscular volume greater than 100 fL), vitamin B12 levels less than 220 pg/mL (normal range 220 to 1130 pg/mL), recovery from anaemia after treatment with intramuscular vitamin B12, hypochlorhydria/achlorhydria to pentagastrin stimulation, and histologic confirmation of gastric body mucosa atrophy.

Conducted in Italy

Table 3Summary of non-randomised studies included in the evidence review

Study

Intervention and comparison

Population

Outcomes

Comments

Armbrecht 1990¹

Gastroscopy N=12

Histories of clinical symptoms taken and upper gastrointestinal endoscopies performed at a mean interval of 14 months

Versus

No monitoring N=61

No further information

Patients recruited to a previous study on gastroscopic screening in pernicious anaemia.

Population strata (previous gastric surgery): not reported

Age: 20-79 years (median 63 years at first investigation)

35 women, 44 men

At mean 6.4 years:

Mortality

Gastric carcinoma

Unclear how 12 patients were selected for regular follow up

Conducted in UK

Sjoblom 1988⁶

Gastroscopy N=71

Gastroscopic screening

Versus

No monitoring N=34

Patients who did not attend screening.

Patients with pernicious anaemia, examined as inpatients (1972-1985) or outpatients (1980-1985); aged ≤75 years

Population strata (previous gastric surgery): no information reported

Age: ≤75 years, no further information reported

Sex: no information reported

At mean 7 years:

Gastric carcinoid tumours

Gastric carcinoma

Criteria for diagnosis of PA:

1. Macrocytic anaemia and/or megaloblastic bone marrow and/or subnormal serum levels of vitamin B12.

2. Schilling test showing intrinsic factor deficiency, pentagastrin-fast achlorhydria or histologically verified fundic atrophic gastritis.

Conducted in Finland

Table 4Clinical evidence summary: Gastroscopy at 24 months versus gastroscopy at 48 months (RCT evidence)

Outcomes

№ of participants (studies)

Follow-up

Certainty of the evidence

(GRADE)

Relative effect

(95% CI)

Anticipated absolute effects

Risk with gastroscopy at 48 months

Risk difference with Gastroscopy at 24 months

Carcinoid tumours

(1 RCT)

⊕◯◯◯

Very low^a^,^b^,^c

OR 0.12

(0.00 to 6.20)

50 per 1,000

50 fewer per 1,000

(180 fewer to 80 more)^d

a: High risk of bias due to lack of information reported on the randomisation process, deviations from the intended interventions and missing outcome data

b: Very serious population indirectness due to lack of information reported on previous gastric surgery and not all participants having pernicious anaemia

c: Downgraded by 1 increment if the confidence interval crossed one MID and by 2 increments if the confidence interval crossed two MIDs (0.8 and 1.25 for dichotomous outcomes; 0.5 * median of baseline SD of the intervention and control group for continuous outcomes).

d: Absolute effects calculated using risk difference

Table 5Clinical evidence summary: Gastroscopy (mean every 14 months) versus no monitoring

Outcomes

№ of participants (studies)

Follow-up

Certainty of the evidence

(GRADE)

Relative effect

(95% CI)

Anticipated absolute effects

Risk with no monitoring

Risk difference with Gastroscopy (mean every 14 months)

Mortality

(1 observational study)

⊕◯◯◯

Very low^a^,^b^,^c

RR 0.73

(0.10 to 5.38)

115 per 1,000

31 fewer per 1,000

(103 fewer to 503 more)

Gastric carcinoma

(1 observational study)

⊕◯◯◯

Very low^b^,^d^,^e

RD 0.00

(−0.11 to 0.11)

0 per 1,000

0 fewer per 1,000

(0 fewer to 0 fewer)

a: Very high risk of bias due to confounding and classification of interventions

b: Serious population indirectness due to lack of information reported on previous gastric surgery

c: Downgraded by 1 increment if the confidence interval crossed one MID and by 2 increments if the confidence interval crossed two MIDs (0.8 and 1.25 for dichotomous outcomes; 0.5 * median of baseline SD of the intervention and control group for continuous outcomes)

d: Very high risk of bias due to confounding, classification of interventions and measurement of outcomes

e: Serious imprecision (risk difference and sample size >70<350)

Table 6Clinical evidence summary: Gastroscopic screening versus no monitoring

Outcomes

№ of participants (studies)

Follow-up

Certainty of the evidence

(GRADE)

Relative effect

(95% CI)

Anticipated absolute effects

Risk with no monitoring

Risk difference with Gastroscopic screening

Carcinoid tumours

105

(1 observational study)

⊕◯◯◯

Very low^a^,^b^,^c

OR 4.66

(0.69 to 31.45)

0 per 1,000

70 more per 1,000

(0 fewer to 140 more)^d

Gastric carcinoma

105

(1 observational study)

⊕◯◯◯

Very low^a^,^b^,^e

RD 0.00

(−0.04 to 0.04)

0 per 1,000

0 fewer per 1,000

(40 fewer to 40 more)

a: Very high risk of bias due to confounding, classification of interventions and missing data

b: Serious population indirectness due to lack of information reported on previous gastric surgery

c: Downgraded by 1 increment if the confidence interval crossed one MID and by 2 increments if the confidence interval crossed two MIDs (0.8 and 1.25 for dichotomous outcomes; 0.5 * median of baseline SD of the intervention and control group for continuous outcomes)

d: Absolute effect calculated using risk difference

e: Serious imprecision (risk difference and sample size >70<350)

Table 7Health economic evidence profile: Gastroscopy surveillance for people with pernicious anaemia

Study

Applicability

Limitations

Other comments

Incremental cost^(c)

Incremental effects

Cost effectiveness

Uncertainty

Lahner 2017³ (Italy)

Partially applicable^(a)

Potentially serious limitations^(b)

Simple cost-effectiveness analysis based on cohort from Italy.
Population: Pernicious anaemia
A) without extensive atrophy (n=79)
B) with extensive atrophy (n=23)

Time horizon: 7.5 years (range 4 – 23.4 years)

A): £376

B): £234

A): 2/79

B): 2/23

A): £13,346 per cancer detected

B): £2,692 per cancer detected

No sensitivity analysis reported.

(a): The population included all have pernicious anaemia with an appropriate intervention – gastroscopy as per the protocol. Although the study does not have a UK NHS perspective, the study is conducted in a similar system compared to the NHS and the healthcare perspective used is appropriate. The use of gastroscopy also reflects current practice in the NHS. However, no QALYs reported. Furthermore, discounting is not applied/reported.

(b): QALYs are not reported, and health outcomes are not included. The only cost incorporated is gastroscopy, no other costs such as treatment costs are reported or the costs of further investigations such as biopsy. There was no sensitivity analysis conducted. The sample size was small.

(c): 2017 Italy Euros converted to UK pounds⁵. Cost components incorporated: Gastroscopy.

Table 13Studies excluded from the clinical review

Study	Code [Reason]
Affronti, J and Baillie, J (1994) Gastroscopic follow-up of pernicious anemia patients. Gastrointestinal endoscopy 40(1): 129 [PubMed: 8163130]	- Review article but not a systematic review
Anonymous. (1998) The role of endoscopy in the surveillance of premalignant conditions of the upper gastrointestinal tract. Gastrointestinal Endoscopy 48(6): 663–668 [PubMed: 9852468]	- Guideline
Arvanitakis, C.; Holmes, F.F.; Hearne III, E. (1979) A possible association of pernicious anemia with neoplasia. Oncology 36(3): 127–129 [PubMed: 471423]	- Study design not relevant to this review protocol
Banks, Matthew, Graham, David, Jansen, Marnix et al. (2019) British Society of Gastroenterology guidelines on the diagnosis and management of patients at risk of gastric adenocarcinoma. Gut 68(9): 1545–1575 [PMC free article: PMC6709778] [PubMed: 31278206]	- Guideline
BIRD, R M (1953) Detection of carcinoma of the stomach in patients with pernicious anemia. Southern medical journal 46(5): 434–9 [PubMed: 13056789]	- Review article but not a systematic review
BOON, T H, SCHADE, R O, MIDDLETON, G D et al. (1964) AN ATTEMPT AT PRESYMPTOMATIC DIAGNOSIS OF GASTRIC CARCINOMA IN PERNICIOUS ANAEMIA. Gut 5: 269–70 [PMC free article: PMC1552122] [PubMed: 14178714]	- Study design not relevant to this review protocol
Borch, K (1986) Epidemiologic, clinicopathologic, and economic aspects of gastroscopic screening of patients with pernicious anemia. Scandinavian journal of gastroenterology 21(1): 21–30 [PubMed: 3952448]	- Data not reported in an extractable format or a format that can be analysed
Borch, K and Liedberg, G (1984) Prevalence and incidence of pernicious anemia. An evaluation for gastric screening. Scandinavian journal of gastroenterology 19(2): 154–60 [PubMed: 6719030]	- Study design not relevant to this review protocol
Bresky, G, Mata, A, Llach, J et al. (2003) Endoscopic findings in a biennial follow-up program in patients with pernicious anemia. Hepato-gastroenterology 50(54): 2264–6 [PubMed: 14696513]	- Study design not relevant to this review protocol
Brinton, L A, Gridley, G, Hrubec, Z et al. (1989) Cancer risk following pernicious anaemia. British journal of cancer 59(5): 810–3 [PMC free article: PMC2247229] [PubMed: 2736218]	- Study design not relevant to this review protocol
Elsborg, L; Andersen, D; Bastrup-Madsen, P (1973) Gastrocamera screening in pernicious anaemia. With special reference to the occurrence of gastric polyps and cancer. Scandinavian journal of gastroenterology 8(1): 5–8 [PubMed: 4697066]	- Study design not relevant to this review protocol
Elsborg, L, Andersen, D, Myhere-Jensen, O et al. (1977) Gastric mucosal polyps in pernicious anaemia. Scandinavian journal of gastroenterology 12(1): 49–52 [PubMed: 65002]	- Study design not relevant to this review protocol
Hsing, A W, Hansson, L E, McLaughlin, J K et al. (1993) Pernicious anemia and subsequent cancer. A population-based cohort study. Cancer 71(3): 745–50 [PubMed: 8431855]	- Study design not relevant to this review protocol
Hughes, Jing W, Muegge, Brian D, Tobin, Garry S et al. (2017) HIGH-RISK GASTRIC PATHOLOGY AND PREVALENT AUTOIMMUNE DISEASES IN PATIENTS WITH PERNICIOUS ANEMIA. Endocrine practice : official journal of the American College of Endocrinology and the American Association of Clinical Endocrinologists 23(11): 1297–1303 [PubMed: 29190137]	- Study design not relevant to this review protocol
Jordan, Paul H Jr; Barroso, Alberto; Sweeney, John (2004) Gastric carcinoids in patients with hypergastrinemia. Journal of the American College of Surgeons 199(4): 552–5 [PubMed: 15454137]	- Study design not relevant to this review protocol
Kokkola, A, Sjoblom, S M, Haapiainen, R et al. (1998) The risk of gastric carcinoma and carcinoid tumours in patients with pernicious anaemia. A prospective follow-up study. Scandinavian journal of gastroenterology 33(1): 88–92 [PubMed: 9489914]	- Study design not relevant to this review protocol
Lahner, Edith, Hassan, Cesare, Esposito, Gianluca et al. (2017) Cost of detecting gastric neoplasia by surveillance endoscopy in atrophic gastritis in Italy: A low risk country. Digestive and liver disease : official journal of the Italian Society of Gastroenterology and the Italian Association for the Study of the Liver 49(3): 291–296 [PubMed: 28034664]	- Study design not relevant to this review protocol
Lehtola, J.; Karttunen, T.; Krekela, I. (1985) Gastric carcinoids with minimal or no macroscopic lesion in patients with pernicious anemia. Hepato-Gastroenterology 32(2): 72–76 [PubMed: 4007765]	- Study design not relevant to this review protocol
Schafer, L W, Larson, D E, Melton, L J 3rd et al. (1985) Risk of development of gastric carcinoma in patients with pernicious anemia: a population-based study in Rochester, Minnesota. Mayo Clinic proceedings 60(7): 444–8 [PubMed: 4010342]	- Study design not relevant to this review protocol
Sjoblom, S M; Sipponen, P; Jarvinen, H (1993) Gastroscopic follow up of pernicious anaemia patients. Gut 34(1): 28–32 [PMC free article: PMC1374095] [PubMed: 8432447]	- Duplicate reference
Sjoblom, S.M.; Sipponen, P.; Jarvinen, H. (1993) Gastroscopic follow up of pernicious anaemia patients. Gut 34(1): 28–32 [PMC free article: PMC1374095] [PubMed: 8432447]	- Review article but not a systematic review
Stockbrugger, R W, Menon, G G, Beilby, J O et al. (1983) Gastroscopic screening in 80 patients with pernicious anaemia. Gut 24(12): 1141–7 [PMC free article: PMC1420244] [PubMed: 6642278]	- Study design not relevant to this review protocol
Ye, W and Nyren, O (2003) Risk of cancers of the oesophagus and stomach by histology or subsite in patients hospitalised for pernicious anaemia. Gut 52(7): 938–41 [PMC free article: PMC1773712] [PubMed: 12801947]	- Study design not relevant to this review protocol

Evidence review for monitoring for gastric cancer in people with vitamin B12 deficiency due to autoimmune gastritis

1. Monitoring for gastric cancer in people with vitamin B12 deficiency due to autoimmune gastritis

1.1. Review question

1.1.1. Introduction

1.1.2. Summary of the protocol

1.1.3. Methods and process

1.1.4. Effectiveness evidence

1.1.4.1. Included studies

1.1.4.2. Excluded studies

1.1.5. Summary of studies included in the effectiveness evidence

1.1.6. Summary of the effectiveness evidence

1.1.7. Economic evidence

1.1.7.1. Included studies

1.1.7.2. Excluded studies

1.1.8. Summary of included economic evidence

1.1.9. The committee’s discussion and interpretation of the evidence

1.1.9.1. The outcomes that matter most

1.1.9.2. The quality of the evidence

1.1.9.3. Benefits and harms

1.1.9.4. Cost effectiveness and resource use

Economic evidence

Potential for modelling

Conclusions about cost effectiveness

Recommendations

Resource impact

1.1.10. Recommendations supported by this evidence review

1.1.11. References

Appendices

Appendix A. Review protocols

Appendix B. Literature search strategies

B.1. What monitoring should be offered to people with pernicious anaemia to identify gastric cancer?

Appendix C. Effectiveness evidence study selection

Appendix D. Effectiveness evidence

Appendix E. Forest plots

Appendix F. GRADE tables

Appendix G. Economic evidence study selection

Appendix H. Economic evidence tables

Appendix I. Health economic model

Appendix J. Excluded studies

Clinical studies

Health Economic studies

Appendix K. Research recommendations – full details

K.1. Research recommendation

K.1.1. Why this is important

K.1.2. Rationale for research recommendation

K.1.3. Modified PICO table

Table 1PICO characteristics of review question

Table 2Summary of randomised controlled trials included in the evidence review

Table 3Summary of non-randomised studies included in the evidence review

Table 4Clinical evidence summary: Gastroscopy at 24 months versus gastroscopy at 48 months (RCT evidence)

Table 5Clinical evidence summary: Gastroscopy (mean every 14 months) versus no monitoring

Table 6Clinical evidence summary: Gastroscopic screening versus no monitoring

Table 7Health economic evidence profile: Gastroscopy surveillance for people with pernicious anaemia

Table 13Studies excluded from the clinical review